This study will examine the effects of socio-ecological factors on population-based cancer screening behaviors, specifically for breast, cervix, colorectal and prostate cancers. "Socio-ecological" factors are community-level social and other environmental characteristics that may influence individuals' utilization of cancer screening. Most importantly, this investigation focuses on identifying which socio-ecological conditions contribute to racial and ethnic differences in cancer screening. The specific aims of this study are (1) to gather and link data at the individual-level, zip code-level and county-level that are relevant to cancer screening access and utilization; (2) with these data, to construct a multivariate multilevel model that predicts cancer screening utilization; and (3) use this multivariate multilevel model framework to measure racial and ethnic differentials in access and utilization of cancer screening tests by interacting individual race/ethnicity variables with selected socio-ecological variables. In this study, I will examine three domains of socio-ecological factors to assess their impact on individuals' utilization of cancer screening: (1) social resources, (2) health services, and (3) health insurance markets. In this multilevel framework, level I is measured at the individual level; it includes the dependent variable (receipt of a cancer screening test) and all associated individual-level characteristics such as sociodemographic information, health status and health insurance. Levels 2 and 3 reflect socio-ecological factors: level 2 consists of aggregate measures of social resources at the zip code-level, for example the proportion of households below poverty level, the percent of non-citizens, and the proportion of households with children; and level 3 includes county-level or county-group-level characteristics for health services and health insurance markets, for example, the percent of minority primary care physicians, the percent of oncologists and the per capita number of community health clinics. The study will use individual-level data from the California Health Interview Survey (CHIS) and socio-ecological variables drawn from a variety of sources. Combining these managed care, competition variables, with social resources and health services capacity provides a comprehensive evaluation of the health system's mediating effect on promoting cancer screening behavior of minority communities. This study complements and advances the knowledge on individual determinants of cancer screening disparities among racial and ethnic groups. And, in analyzing a comprehensive data set with multiple-levels of information, I can target needed policy changes in social and health market conditions in neighborhoods and in counties to reduce disparities in cancer screening.